NeuralHydrology -- Interpreting LSTMs in Hydrology

Abstract

Despite the huge success of Long Short-Term Memory networks, their applications in environmental sciences are scarce. We argue that one reason is the difficulty to interpret the internals of trained networks. In this study, we look at the application of LSTMs for rainfall-runoff forecasting, one of the central tasks in the field of hydrology, in which the river discharge has to be predicted from meteorological observations. LSTMs are particularly well-suited for this problem since memory cells can represent dynamic reservoirs and storages, which are essential components in state-space modelling approaches of the hydrological system. On basis of two different catchments, one with snow influence and one without, we demonstrate how the trained model can be analyzed and interpreted. In the process, we show that the network internally learns to represent patterns that are consistent with our qualitative understanding of the hydrological system.

arXiv commentPre-print of published book chapter. See journal reference and DOI for more info

References in corpus (5)

• Axiomatic Attribution for Deep Networks
• Explaining NonLinear Classification Decisions with Deep Taylor Decomposition
• Visualizing and Understanding Recurrent Networks
• Learning how to explain neural networks: PatternNet and PatternAttribution
• Beyond Word Importance: Contextual Decomposition to Extract Interactions from LSTMs

Cited by in corpus (8)

• Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications
• Enhancing streamflow forecast and extracting insights using long-short term memory networks with data integration at continental scales
• Toward Explainable AI for Regression Models
• Capabilities of Deep Learning Models on Learning Physical Relationships: Case of Rainfall-Runoff Modeling with LSTM
• Time Series Predictions in Unmonitored Sites: A Survey of Machine Learning Techniques in Water Resources
• Disentangled Explanations of Neural Network Predictions by Finding Relevant Subspaces
• Inferring the past: a combined CNN-LSTM deep learning framework to fuse satellites for historical inundation mapping
• MC-LSTM: Mass-Conserving LSTM